In the literature, two basic approaches are proposed for the design of the durability of Reinforced Concrete (RC) structures in aggressive ACCEPTED MANUSCRIPT 3 environments [1]: a prescriptive approach and a performance-based approach. The prescriptive approach is primarily based on the acquired experience in the durability performance of existing RC structures. Because experience is generally insufficient to allow for the quantitative requirements, most of the requirements for durability are formulated in a qualitative and empirical way. In the case of reinforcing steel corrosion due to carbonation or chlorides, the prescriptive approach defines an exposure class and subsequent prescriptions including (i) concrete composition (a maximum water-to-cement ratio, a minimum cement content and a cement type); (ii) a minimum 28-day compressive strength of the concrete; and (iii) a minimum concrete cover depth for service life design [2] [3]. The key feature of the performance-based approach is to assess relevant concrete material properties using some relevant test methods or service life prediction models. This approach can be used to formulate requirements as regards material properties and structure dimensions. In the case of corrosion of reinforcing steel due to carbonation [4] [1] [5] or chlorides [6] [7] [8], the estimation of the deterioration evolution depending on expected influential parameters is mostly performed by applying a probabilistic approach. This estimation makes it possible to formulate requirements for the structural responses depending on the service life design [3]. Then, durability design can be completed in two ways: (i) using a fully probabilistic method, for which the concrete cover depth and the diffusion coefficient of CO2 or chlorides are usually considered as main probabilistic design parameters for the required service life design and the reliability level [4] [1] [5] [6] [7] [8]; and (ii) using the partial factor method to determine the characteristic values and the partial factors for the design parameters [4] [1] [7].